Biocides do what they are supposed to do: control harmful or undesirable organisms. Depending on the nature of their application, however, they can enter surface water (and possibly drinking water) via different pathways. Here they are harmful to desirable organisms. What do we know about the emission pathways of biocides and the risks they pose to people and nature? And how do we develop efficient and effective monitoring strategies?
Biocide literally means life-killing. Legislation defines biocides as products containing one or more active substances to control (destroy, repel or render harmless) harmful or unwanted organisms. Therefore, they typically contain toxic substances. This makes them potentially harmful to organisms that do not need to be controlled, even after emission into the (water) environment.
In the Knowledge Impulse for Water Quality (KIWK), the national government, provinces, water boards, drinking water utilities and knowledge institutes work together to gain more insight into the quality of groundwater and surface water and the factors that influence this quality. The KIWK Ketenverkenner project has, amongst others, aggregated knowledge about biocides and their effects on water quality. The study which we describe in this article, investigated what is known about the presence of biocides in the water system and the risks associated with this presence, as well as how this knowledge can be used in developing monitoring approaches.
Emission pathways are divers
By monitoring chemicals in the water, we know whether emissions are causing problems for water quality. But biocides are hardly monitored, possibly because little is known about the extent of use and emission pathways. Based on the nature of their application, the European Biocides Regulation distinguishes four main groups of biocidal products: disinfectants, preservatives, pest control agents and others. In total, the four groups comprise 22 product types [1]. These include antifouling agents for ships, preservatives in building materials, insect repellents, agents to prevent biofilm in cooling towers or algae in artificial turf.
As the applications of biocides are very diverse, so are the routes of spreading to water. Disinfectants often end up in water via the sewage treatment plant. This also applies to preservatives, but these can also end up in surface water directly from, for example, building materials or wood or in groundwater via the soil. By discharge of cooling water or the use of antifoulants, direct emission to surface water is possible. Pest control agents can reach the water through all these routes.
Model-based regulation
The active substances in biocidal products must have received European approval or be in the review programme. Subsequently, the biocides are tested before they can be sold on the Dutch market. This is done by the Dutch Board for the Authorisation of Plant Protection Products and Biocides (Ctgb). The review examines whether the product is sufficiently effective and has no significant adverse effect on humans and the environment. An authorised biocidal product has binding instructions for use. It states where and how it may be used, and how waste and treated material should be dealt with after use.
The assessment of biocidal products is based on estimations and models. If actual use deviates or the models do not describe all situations well, differences arise between assumed and actual emissions and environmental concentrations. Only by monitoring in the water system can these differences come to light. However, there are around 270 substances in use as biocides. It is impossible to measure them all everywhere.
Biocide measurements in the water system
Which biocides are currently being monitored in surface and groundwater? An analysis of monitoring data from the Water Quality Portal (data from 2019) shows that 18% of the authorised active substances are actually being measured. For groundwater (with fewer measurements in the Water Quality Portal and therefore analysed for the period 2000-2018), this percentage is 15%. Respectively 13% and 10% of the permitted substances are actually found (Figure 1).
Figure 1. Percentage of active substances in biocidal products monitored and detected in surface water (2019) and in groundwater (2000-2018) Source: Water quality portal.
At most groundwater locations, all measured substances were below the reporting limit. In surface water, biocides are widespread: they were found above the reporting limit at two-thirds of the sites. One third of the monitored substances were found (once or several times) above the detection limit, both in groundwater as well as in surface water [2]. (Signal value: when the concentration of a new (emerging) substance in intake water for drinking water production structurally exceeds 0.1 µg/l for 3 years, the health risks are investigated).
Drawing up measurement strategies for biocides
Which substances should be included in a monitoring programme? Ideally this decision is based on 1) the probability of their occurrence and 2) the impact they have on water quality. The probability of detection is determined by the substance properties, the selection of the monitoring sites and the timing of the monitoring. Looking at the current data, it appears that targeted monitoring for biocides is either lacking or not performed in the most relevant locations. Measured data on biocides are almost always by-catch of measurement programmes aimed at other substance groups, such as plant protection products [2]. In addition, many active substances in biocidal products also have other uses; therefore, it is unclear whether a substance found in a biocidal product originates from its use as a biocide. To be able to say something about the risks that substances pose to people and nature, more measurement data is needed.
Developing robust and sensitive measurement methods and integrating them into monitoring programmes is not always possible or easy. Therefore, in order to draw up an effective monitoring strategy, a good selection of substances to be monitored is necessary [6]. The Ketenverkenner project has developed a decision tree for this purpose (Figure 3) [4].
The decision tree distinguishes four steps for the selection. In step 1, the likelihood of exposure to (finding) a biocide is determined for the local water system. This is based on knowledge of use, application, substance properties (see box) and hydrology. In step 2, the potential impact on drinking water quality or ecological surface water quality is determined. Together these determine the risk of the substance: that is step 3. In step 4, for the relevant (‘prioritised’) substances, it is then examined whether a suitable measurement method is available (with a sufficiently low detection limit) or whether it should be developed. When a method is available (or developed), inventory measurements are taken at selected locations. Based on the results, it will be decided in step 5 whether the substances in question will be included in regular monitoring.
Figure 3: Decision tree for the inclusion or exclusion of a substance in a measurement strategy.
In order to maintain an adequate monitoring programme, it is essential to continue surveying. This is to avoid measuring only what we already know and keeping new substances out of sight – the so-called confirmation bias – and to test assumptions about exposure. If a substance is structurally not found, monitoring can be discontinued [7]. Without such continuous evaluation, the monitoring programme loses its effectiveness in the long run.
Authors
Thomas ter Laak, Tessa Pronk, Joep van den Broeke (KWR), Joke Wezenbeek, Els Smit (RIVM), Ivo Roessink, Sanne van den Berg, Bas Buddendorf (Wageningen Environmental Research)
Sources
- Ctgb (2022), website consulted on 11-04-2022. http://www.ctgb.nl/biociden
- Pronk, T.E., Wezenbeek, J., & Buddendorf, B. (2022). Biocides Excel Table (Version v2) [Data set]. https://doi.org/10.5281/zenodo.6361682
- Wezenbeek, J., Roessink, I., van den Berg, S., ter Laak, T., Pronk, T. (2021). Deltafact Biocides. STOWA: http://www.stowa.nl/deltafacts/waterkwaliteit/kennisimpuls-waterkwaliteit/biociden
- Pronk, T.E., Roessink, I., Smit, E. (2022). MEETING STRATEGY BIOCIDES Considerations and criteria, STOWA report no. 2022-07.
- Reemtsma, T.; Berger, U.; Arp, H. P. H.; Gallard, H.; Knepper, T. P.; Neumann, M.; Benito Quintana, J.; Voogt, P. (2016). Mind the gap: persistent and mobile organic compounds – water contaminants that slip though. Environmental Science & Technology 50(19), 10308 – 10315.
- Pohl, K., et al. (2015). Environmental monitoring of biocides in Europe – compartment-specific strategies. Workshop Report (June 25-26, 2015 in Berlin).
- von der Ohe, P. C. and V. Dulio (2013). NORMAN Prioritisation framework for emerging substances. ISBN: 978-2-9545254-0-2.
—————————————————————————————————————–
Substance properties of biocidal products
In the KIWK project Ketenverkenner, the substance properties of biocides were collected [3] and linked to the monitoring data from the Water Quality Portal (Figure 2). Biocides that are poorly biodegradable appear to be over-represented in the measurements [4], while highly volatile biocides (with high vapour pressure; VP) are hardly observed at all. This can be explained by the fact that easily biodegradable and volatile biocides disappear quickly. Surprisingly, substances that strongly adsorb to soil or sediment (with a high sorption coefficient of organic carbon, Koc) are found in high concentrations in water. In this it’s certainly a factor that these substances are over-represented in environmental research because their isolation and analysis is easier [5].
Figure 2. The occurrence (frequency) of biocides with a certain substance property in the water system in relation to the total occurrence of biocides with those substance properties. Displayed properties: biodegradability, vapour pressure (log VP) and organic carbon sorption coefficient (log Koc).
Of the 116 biocides that, based on their properties, could have meaningful emissions to the water system, for 83 there is no known measurement method [2]. These substances are therefore not monitored, even if they are or may be relevant to water quality. In order to get a better picture, it is important to develop measuring methods and to actually start monitoring in a targeted manner.
